pending-medicare-mcp
Read-only MCP server providing a validated public preview of behavioral-health NPIs newly present in CMS pending first-time Medicare enrollment files, including state/specialty counts and official methodology.
README
Pending Medicare Enrollment Data MCP server
Read the current validated public preview of behavioral-health NPIs newly present in CMS pending first-time Medicare enrollment files. The server is remote, read-only, unauthenticated, and dependency-free for clients.
Connect
Claude Code:
claude mcp add --transport http pending-medicare https://actablesite.com/mcp/pending-medicare
For another MCP client, add a Streamable HTTP server using https://actablesite.com/mcp/pending-medicare. No API key or authorization header is required.
Or inspect and connect through the automatically indexed Glama connector. Glama currently reports the endpoint healthy and exposes its tool schema without requiring a manual submission.
Tool
get_pending_medicare_behavioral_health_preview returns:
- the current 10-row public sample, optionally filtered by up to 10 two-letter state or territory codes;
- current and prior CMS snapshot dates;
- the validated national record count, state counts, and specialty counts;
- official methodology, the public source receipt, and material limitations; and
- an optional $12 full-edition handoff to the buyer-funded Apify Actor.
The tool cannot create an Apify token, start a run, open checkout, contact an applicant, send data elsewhere, or purchase anything. The full edition remains a separate intentional event; buyer-paid Apify platform usage is separate.
Interpretation boundary
Pending means a first-time Medicare enrollment application is still under Medicare Administrative Contractor review. It does not prove approval, enrollment, billing privileges, credentialing, licensure, a new practice, availability, interest, or buying intent. Public NPPES contact fields may be old, shared, or operational. Verify every record before relying on it.
Verify
npm test
npm run verify:live
The live verifier initializes the endpoint, confirms the one-tool read-only inventory, retrieves the free preview, and checks its source dates, counts, price boundary, and limitations. It does not start a paid run or create a purchase.
Registry
The server is published through GitHub OIDC as io.github.unitedideas/pending-medicare-behavioral-health-provider-enrollment. The release workflow needs no publishing secret.
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。